Cognitive dietary restraint (CDR) is the perception of limiting food intake to achieve/maintain a perceived ideal body weight. The objective of this 2‐yr prospective study was to determine if women with higher CDR have more subclinical ovulation disturbances (SOD; anovulation or luteal phase <10 d), elevated cortisol excretion and less positive changes in areal bone mineral density over two years (ΔaBMD). Participants were 123 healthy, non‐obese, regularly menstruating women aged 19–35. Annually participants completed questionnaires (Three Factor Eating Questionnaire, physical activity, general stress, dietary intake), anthropometrics and 24‐hr urine collections for measurement of 24‐hr urinary free cortisol (UFC). At baseline and final follow‐up, dual energy x‐ray absorptiometry scans measured ΔaBMD at the lumbar spine (L1‐4), hip and total body. Quantitative waking temperature was used to determine SOD frequency. SOD correlated positively with CDR and inversely with hip ΔaBMD but not with UFC. CDR and UFC were not associated with each other or ΔaBMD. Women with higher CDR had more SOD. Women with more SOD had less positive ΔaBMD and higher CDR. Physical activity, general stress, body mass and energy intake did not explain differences by CDR or SOD. In conclusion, CDR was the only study variable associated with SOD which was associated with less positive ΔaBMD. Canadian Institutes of Health Research 79563. Grant Funding Source : Canadian Institute of Health Research
cognitive dietary restraint ovulation disturbances cortisol bone density premenopausal, Prior JC subclinical ovulation disturbances bone mineral density, anovulation short luteal phase areal bone mineral density change, Bedford Prior Barr dietary restraint ovulation bone FASEB, basal waking temperature ovulation detection prospective study, dietary restraint eating behavior menstrual cycle disruption, premenopausal women hip spine bone density two year change, subclinical anovulation cortisol excretion body mass, luteal phase length less than 10 days bone loss, Three Factor Eating Questionnaire restraint score ovulatory disturbances
DOI 10.1096/fasebj.24.1_supplement.561.12 10.1096/fasebj.24.1_supplement.561.12
Cite this article
Bedford, J. L., Prior, J., & Barr, S. I. (2010). Relationships among cognitive dietary restraint, subclinical ovulation disturbances, cortisol and 2‐yr change in bone density in premenopausal women. The FASEB Journal, 24(S1). https://doi.org/10.1096/fasebj.24.1_supplement.561.12
Bedford JL, Prior J, Barr SI. Relationships among cognitive dietary restraint, subclinical ovulation disturbances, cortisol and 2‐yr change in bone density in premenopausal women. The FASEB Journal. 2010;24(S1). doi:10.1096/fasebj.24.1_supplement.561.12
Bedford, J. L., et al. "Relationships among cognitive dietary restraint, subclinical ovulation disturbances, cortisol and 2‐yr change in bone density in premenopausal women." The FASEB Journal, vol. 24, no. S1, 2010.
Related articles
Menstrual CycleSubclinical AnovulationOvulatory Disturbance EffectsStress and Ovulation
Cross-sectional studies have found associations among elevated cognitive dietary restraint (CDR), increased ovulatory disturbances, and lower bone mass, possibly mediated by cortisol. To determine whether healthy young women with higher CDR have more menstrual cycles with subclinical ovulatory disturbances (SOD), elevated 24-h urinary free cortisol (UFC), and less positive 2-yr areal bone mineral density change (Delta-aBMD). DESIGN, SETTING, We conducted a 2-yr longitudinal study of 123 healthy, community-dwelling, nonobese, regularly menstruating women aged 19-35 yr. Key variables were Three Factor Eating Questionnaire Restraint score, percent of cycles with anvoluation and/or luteal phase length <10 d (%SOD), UFC, and Delta-aBMD at the lumbar spine (L1-L4), total hip, and whole body. Anthropometrics, general stress, physical activity, and energy intake were measured. Adjusting for potential confounders, differences were examined by general linear modeling using median split of CDR score and %SOD. Women with higher CDR had higher %SOD (56 vs. 34%, P < 0.001) and higher UFC (28.0 vs. 24.0 microg/d, P = 0.021). Delta-aBMD did not differ by CDR. Women with higher %SOD had less positive Delta-aBMD at L1-L4 (0.7 vs. 1.9%, P = 0.034) and hip (-0.6 vs. 0.9%, P = 0.001), and higher CDR score (8.7 vs. 7.1, P = 0.04). Physical activity, general stress, body mass index, and energy intake did not explain differences by CDR or %SOD. UFC was not associated with %SOD or Delta-aBMD. Women with more frequent SOD reported higher CDR and experienced less positive Delta-aBMD. Although women with higher CDR had higher UFC, the mechanism linking CDR, SOD, and aBMD is not clear.
Menstrual CycleSelf-Worth and MoodBasal Body TemperatureProspective Ovulation Cohort
Open Access
Subclinical ovulatory disturbances (anovulation or short luteal phases within normal-length menstrual cycles) indicate lower progesterone-to-estrogen levels. Given that progesterone plays a bone formation role, subclinical ovulatory disturbances may be associated with bone loss or less than expected bone gain. Our purpose was to perform a meta-analysis of prospective studies in healthy premenopausal women to determine the overall relationship of subclinical ovulatory disturbances to change in bone mineral density. Two reviewers independently identified from serial literature searches 6 studies a 2-year study in 114 young adult women, 2006-2009, Vancouver, Canada; a 2-year study in 189 premenopausal women, 2000-2005, Toronto, Canada; a single-cycle study in 14 young women, 1996-1997, Melbourne, Australia; an 18-month study in 53 women, 1990-1995, Santa Clara, California; a 4-year study in 27 women, 1988-1995, Vancouver, Canada; and a 1-year study in 66 women, 1985-1988, Vancouver, Canada. This meta-analysis included a combined sample size of 473 observations in 436 premenopausal women studied over 1-4 years and aged 14-47 years. The percentage of women with ovulatory disturbances varied significantly from 13% to 82%. Women with more frequent ovulatory disturbances had more negative percentage changes in spine bone mineral density (weighted mean difference = -0.86; P = 0.040) for random-effects analysis. There was significant heterogeneity among these 6 studies (I(2) = 80%). In summary, these data show that regularly menstruating women with more frequent ovulatory disturbances experience more negative changes in bone (approximately -0.9% per year). These cycles with silent estrogen/progesterone imbalance may be clinically important.
Fertility AwarenessBasal Body TemperatureTemperature Analysis MethodsProgesterone Confirmation Methods
Jennifer L Bedford et al., 2009·Eur J Obstet Gynecol Reprod Biol
To assess computerised least-squares analysis of quantitative basal temperature (LS-BT) against urinary pregnanediol glucuronide (PdG) as an indirect measure of ovulation, and to evaluate the stability of LS-QBT to wake-time variation. Cross-sectional study of 40 healthy, normal-weight, regularly menstruating women aged 19-34. Participants recorded basal temperature and collected first void urine daily for one complete menstrual cycle. Evidence of luteal activity (ELA), an indirect ovulation indicator, was assessed using Kassam's PdG algorithm, which identifies a sustained 3-day PdG rise, and the LS-QBT algorithm, by determining whether the temperature curve is significantly biphasic. Cycles were classified as ELA(+) or ELA(-). We explored the need to pre-screen for wake-time variations by (A) all recorded temperatures, (B) wake-time adjusted temperatures, (C) temperatures within 2h of average wake-time, and (D) expert reviewed temperatures. Relative to PdG, classification of cycles as ELA(+) was 35 of 36 for LS-QBT methods A and B, 33 of 34 (method C) and 30 of 31 (method D). Classification of cycles as ELA(-) was 1 of 4 (methods A and B) and 0 of 3 (methods C and D). Positive predictive value was 92% for methods A-C and 91% for method D. Negative predictive value was 50% for methods A and B and 0% for methods C and D. Overall accuracy was 90% for methods A and B, 89% for method C and 88% for method D. The day of a significant temperature increase by LS-QBT and the first day of a sustained PdG rise were correlated (r=0.803, 0.741, 0.651, 0.747 for methods A-D, respectively, all p<0.001). LS-QBT showed excellent detection of ELA(+) cycles (sensitivity, positive predictive value) but poor detection of ELA(-) cycles (specificity, negative predictive value) relative to urinary PdG. Correlations between the methods and overall accuracy were good and similar for all analyses. Findings suggest that LS-QBT is robust to wake-time variability and that expert interpretation is unnecessary. This method shows promise for use as an epidemiological tool to document cyclic progesterone increase. Further validation relative to daily transvaginal ultrasound is required.